Tutorials

Tutorials

No.Date & TimeTutorial TitleOrganizer
1Oct. 27 (Tue)
09:00-11:00
Image-Based 3D Model Reconstruction: From Multi-View Geometry to Neural RenderingProf. Soohwan Song
(Dongguk University)
2Oct. 27 (Tue)
09:00-11:20
Fun and Foundations of Swarm Systems and ControlProf. Masaki Ogura
(Hiroshima University)
3Oct. 27 (Tue)
09:00-12:10
Dynamically Embedded Model Predictive Control: Real-Time Operation, Constraint Satisfaction, and StabilityProf. Hyungbo Shim
(Seoul National University)
4Oct. 27 (Tue)
13:00-15:50
Learning-Based Control for Automotive Systems: From Semi-Parametric Dynamics Modeling to Online Policy ImprovementProf. Heejin Ahn (KAIST)
5Oct. 27 (Tue)
14:00-17:00
From AI & Robotics Research to Railway DeploymentDr. Claire Nicodeme (ALSTOM)

Tutorial 1

Image-Based 3D Model Reconstruction: From Multi-View Geometry to Neural Rendering

OrganizerProf. Soohwan Song (Dongguk University)
Date & TimeOctober 27 (Tue) 09:00-11:00
VenueHotel Emisia Sapporo Room T2 (Cattleya), 3F

Overview

TimeProgramSpeaker
09:00 – 09:50Multi-View Stereo Theory and Practice: COLMAP / DUST3R / VGGT

Prof. Soohwan Song
(Dongguk University)

09:55 – 10:05Break
10:05 – 11:003D Gaussian Splatting Theory and Practice, and Case Studies in Diffusion-Based 3D Generation

Abstract

This tutorial provides a comprehensive introduction to state-of-the-art techniques for photorealistic 3D model reconstruction from multi-view images. It begins with the core principles of Multi-View Stereo (MVS), a foundational technique in image-based 3D reconstruction, and then guides participants through hands-on exercises using widely adopted open-source tools such as COLMAP, DUST3R, and VGGT. The tutorial further introduces 3D Gaussian Splatting (3DGS), a rapidly advancing neural rendering technique, with practical sessions focused on building high-quality 3D rendering models. Finally, it discusses recent advances that combine generative AI, particularly diffusion models, with 3D modeling for the creation of novel 3D assets.

Presenter Biography

Prof. Soohwan Song
Assistant Professor
Department of Computer Science and Artificial Intelligence,
Dongguk UniversityTopics of recent interest include
– Physical AI
– 3D Vision
– Multi-Robot System
Homepage: https://sites.google.com/view/smrlab/

Tutorial 2

Fun and Foundations of Swarm Systems and Control

OrganizerProf. Masaki Ogura (Hiroshima University)
Date & TimeOctober 27 (Tue) 09:00-11:20
VenueHotel Emisia Sapporo Room T3 (Azalea), 3F

Overview

TimeProgramSpeaker
09:00 – 09:05Opening & IntroductionOrganizer
09:05 – 09:35Understanding and Control of Fish SchoolingProf. Naoki Wakamiya
(The University of Osaka)
09:35 – 09:50Shepherding Dynamics: A Testbed for Nonlinear and Information-Limited Multi-Agent ControlProf. Masaki Ogura
(Hiroshima University)
09:50 – 10:15AI Learning Viewed as Swarm ControlProf. Shigeru Shinomoto
(Ritsumeikan University)
10:15 – 10:20Break 
10:20 – 10:50Swarm Control of Biohybrid Robots: Challenges and OpportunitiesProf. Yang Bai
(Hiroshima University)
10:50 – 11:20From Early Warning to Re-stabilization: Dynamical Network Markers for Critical Transition ControlDr. Hampei Sasahara
(The University of Tokyo)

Lecture 1

Understanding and control of fish schooling
This tutorial introduces recent advances in understanding and controlling a biological collective behavior known as fish schooling. Many fish species spontaneously self-organize themselves to swim in coordination like a single entity, providing benefits such as energy efficiency, improved foraging, and enhanced predator avoidance. Their collective behavior emerges without centralized control: individual fish rely on simple behavioral rules based on local information and respond to their surrounding environment including neighbor fishes. This distributed nature of fish schooling offers valuable insights for the design of scalable, adaptive, and robust multi-agent systems. In particular, control through environmental modulation, rather than precise manipulation of individual agents, enables emergence of desired global outcomes from local interactions and environmental responses. The talk will present examples of mathematical modeling of schooling behavior of living fishes and their validations.

Presenter Biography

Prof. Naoki Wakamiya
Graduate School of Information Science and Technology
The University of Osaka
https://www-waka.ist.osaka-u.ac.jp/index.php/en
Topics of recent interest include:
Bio-inspired ICT, Self-organization, Biological modeling,
multi-agent systems

Lecture 2

Shepherding Dynamics: A Testbed for Nonlinear and Information-Limited Multi-Agent Control
The shepherding problem focuses on guiding a collective of swarm agents using one or more external agents (“sheepdogs”). Since its introduction, shepherding has become a foundational model in swarm control due to its simplicity, universality, and applicability. Beyond its biological origin, shepherding dynamics naturally serve as a testbed for nonlinear and information-limited multi-agent control, where classical linear or convex tools cannot be directly applied. This talk reviews the early development of shepherding and highlights recent advances in handling heterogeneity within swarms, coordinating multiple shepherds under limited or no communication, and performing shepherding using restricted sensing such as bearing-only measurements. These developments provide insights that may inform the design of robust and scalable control strategies for multi-agent systems.

Presenter Biography

Prof. Masaki Ogura
Graduate School of Advanced Science and Engineering
Hiroshima University
https://csslab.jp/en/
Topics of recent interest include:
Swarm control, multi-agent systems, complex systems, artificial intelligence

Lecture 3

AI Learning Viewed as Swarm Control
Following the success of control theories based on linear modelling, recent developments have emerged that address the control of complex nonlinear systems for which accurate modelling is difficult. Swarm control is one such theory. Theoreticians seem to dream of controlling complex systems with many degrees of freedom using systems with fewer degrees of freedom. The AI learning process resembles swarm control because the number of free parameters (~ one trillion) far exceeds the amount of controllable training data (~ one billion). While humans cannot perfectly control or comprehend AIs, they appear to function reasonably well. In my talk, I will introduce some theories that attempt to explain why this is the case. Similarly to AI learning, swarm control might be possible if we don’t try to control everything down to the last detail.

Presenter Biography

Prof. Shigeru Shinomoto
Shigeru Shinomoto studied physics and received his Ph.D. from Tohoku University. After moving to Kyoto University, he began collaborating with Yoshiki Kuramoto on cooperative phenomena in nonlinear oscillators and with Shun-ichi Amari on computational learning theory. He worked at Kyoto University for over 35 years, primarily developing analytical tools for neuroscience data. Since retiring from Kyoto University in 2021, he has continued working as a visiting researcher at Kyoto University, the ATR Institute, Ritsumeikan University, the University of Hyogo, and Hiroshima University.
https://sites.google.com/view/shigerushinomoto/

Lecture 4

Swarm Control of Biohybrid Robots: Challenges and Opportunities
Biohybrid robots, or cyborg systems, combine living organisms with artificial sensing, computation, and stimulation devices. They can be programmed and controlled like robots, while retaining the mobility, energy efficiency, adaptability, and robustness of biological organisms. These properties make cyborg systems promising for operation in complex and unstructured environments. While most existing studies focus on single-agent control, many real-world applications require multiple cyborg agents to work together. Swarm control of biohybrid robots involves two fundamental challenges: biological agents may respond differently to the same stimulation, and electrical stimulation should be minimized to preserve their natural autonomy and capabilities. This talk presents recent efforts toward a general swarm control framework for cyborg systems. The central idea is to coordinate biological agents without treating them as fully controllable machines. Instead, biohybrid swarm control should make use of the natural behavior of living agents and guide the swarm only when necessary. Experimental results on cyborg robot swarms demonstrate that this framework can support diverse coordination tasks, including navigation, transport, and formation control. Overall, this talk highlights a shift from enforcing biological agents to coordinating with them, opening new opportunities for scalable, adaptive, and resilient swarm intelligence in biohybrid robotic systems.

Presenter Biography

Prof. Yang Bai
Specially Appointed Associate Professor
Graduate School of Advanced Science and Engineering
Hiroshima University
https://scholar.google.com/citations?user=_PCUWNYAAAAJ
Topics of recent interest include:
Complex systems, cyborg systems, safe learning-based control, swarm intelligence

Lecture 4

From Early Warning to Re-stabilization: Dynamical Network Markers for Critical Transition Control
Critical transitions are abrupt changes in system behavior that arise in a wide range of complex networked systems, including disease onset in biological networks and cascading failures in power systems. For high-dimensional systems, however, model-based stabilization is often hindered by the limited availability of data, such as high-dimensional low-sample-size (HDLSS) datasets and snapshot observations, making conventional system identification challenging. This tutorial presents a dynamical network marker (DNM)-based framework that links early warning signals of critical transitions to control-oriented intervention. By exploiting dynamical information embedded in DNM structures, the framework enables the construction of dynamical models around critical regimes and the design of re-stabilization strategies via pole shifting. The tutorial introduces the underlying theory, methodological foundations, and illustrative examples, demonstrating how DNM-based approaches can support re-stabilization of complex networked systems before critical transitions occur.

Presenter Biography

Dr. Hampei Sasahara
Lecturer
Graduate School of Information Science and Technology
The University of Tokyo
https://hampei.net/index_en
Topics of recent interest include:
– Data-driven Control
– Control System Security
– Autonomous Cybersecurity Agent
– Control of Large-scale Systems

Tutorial 3

Dynamically Embedded Model Predictive Control: Real-Time Operation, Constraint Satisfaction, and Stability

OrganizerProf. Hyungbo Shim (Seoul National University)
Date & TimeOctober 27 (Tue) 09:00-12:10
VenueHotel Emisia Sapporo Room T4 (Lilac), 3F

Overview

TimeProgramSpeaker
09:00 – 09:10Opening & Tutorial OverviewProf. Hyungbo Shim
(Seoul National University)
09:10 – 09:30Motivations from Vehicle Control and Practical ConsiderationsDr. Jinsung Kim
(Hyundai Motor Company)
09:30 – 10:00Motivation & Problem FormulationProf. Jin Gyu Lee
(Seoul National University)
10:00 – 10:10Break 
10:10 – 10:40Dynamically Embedded MPC for Real-Time OperationProf. Hyungbo Shim
(Seoul National University)
10:40 – 11:10Constrained Optimization from a Control PerspectiveYong Joo Do
(Seoul National University)
11:10 – 11:20Break 
11:20 – 11:50Soft-Constrained MPC: Feasibility and StabilityHyeonyeong Jang
(Seoul National University)
11:50 – 12:10Constraint Satisfaction under Inexact OptimizationHyungjo Byun
(Seoul National University)

Lecture 1

Motivations from Vehicle Control and Practical Considerations

Advanced vehicle control requires coordinated operation of multiple actuators under tight computational budgets and strict timing requirements. These practical demands motivate dynamically embedded predictive control as a lightweight approach that evolves control solutions through embedded dynamics with limited computation per update. This talk discusses its motivation, implementation considerations, and potential applications to coordinated vehicle motion control on embedded automotive platforms.

Presenter Biography

Dr. Jinsung Kim
Jinsung Kim received the B.S. degree in Mechanical and Automotive Engineering from Kookmin University in 2007 and the M.S. and Ph.D. degrees in Mechanical Engineering from the Korea Advanced Institute of Science and Technology (KAIST) in 2009 and 2013, respectively. He joined Hyundai Motor Company in 2013, where he works on control-software development for electrified-vehicle propulsion systems. From 2019 to 2020, he was a visiting scholar at the University of Pennsylvania. His research interests include observer design, data-driven control, and optimal control of electric vehicles

Lecture 2

Motivation & Problem Formulation

This talk motivates the need for reliable real-time implementation of constrained Model Predictive Control (MPC). Conventional MPC assumes its optimal control problem stays feasible and is solved to sufficient accuracy within every sampling interval. In fast or complex systems, however, limited computation can produce inexact control inputs, while abrupt reference changes, disturbances, and model mismatch can cause infeasibility and interrupt the control update. Based on these challenges, the tutorial focuses on three objectives: continuous generation of control inputs in real time, systematic management of feasibility and constraint satisfaction, and closed-loop stability guarantees. It concludes by showing how constrained optimization from a feedback-control perspective, soft constraints, dynamically embedded MPC, reference governance, and constraint tightening address these objectives.

Presenter Biography

Prof. Jin Gyu Lee

Jin Gyu Lee received the B.S. and Ph.D. degrees from Seoul National University. He subsequently held postdoctoral positions at the University of Cambridge and Imperial College London and joined Inria in Lille, France, in 2022. Since 2024, he has been with Seoul National University. His research interests include multi-agent systems, observer design, secure control, nonlinear oscillators, distributed optimization, adaptive control, neuronal networks, and neuromorphic engineering.

Homepage: https://jingyulee92.wordpress.com

Lecture 3

Dynamically Embedded MPC for Real-Time Operation

This talk presents dynamically embedded MPC, in which an evolving estimate of the solution to an optimal control problem is embedded in the internal state of a continuous-time dynamic controller running in parallel with the plant. Starting from the MPC formulation, the Lagrangian and Karush-Kuhn-Tucker residuals are used to construct projected primal-dual optimization dynamics, allowing continuous control-input generation without waiting for exact solver convergence at each sampling instant. The coupled plant-optimizer system is then analyzed using input-to-state stability and a small-gain argument, which clarifies why the optimization dynamics must be sufficiently fast relative to the plant. The talk also provides a step-by-step procedure for constructing the dynamic controller from the optimal control problem and discusses design choices including initialization and optimizer gain.

Presenter Biography

Prof. Hyungbo Shim
Hyungbo Shim received the B.S., M.S., and Ph.D. degrees from Seoul National University and held a postdoctoral position at the University of California, Santa Barbara. After joining Hanyang University in 2002, he moved to Seoul National University in 2003. He has served in editorial roles for major control journals and is the director of the Engineering Research Center for Advanced Control and Instrumentation. His research interests include nonlinear-system stability, observer and disturbance-observer design, secure control systems, and synchronization.
Homepage: http://hshim.wordpress.com

Lecture 4

Constrained Optimization from a Control Perspective

This talk introduces Controlled Multipliers Optimization (CMO), a framework that interprets constrained optimization from a control perspective. In CMO, Lagrange multipliers are viewed not only as dual variables but also as control inputs acting on continuous-time optimization dynamics, while constraint residuals are treated as outputs to be regulated. Starting from equality-constrained optimization, the talk explains how proportional-integral feedback and feedback linearization can be used to construct multiplier dynamics that drive constraint residuals to zero and guide the optimization variables toward stationary points. It also discusses the relationship between CMO and primal-dual gradient dynamics, including the effect of proportional gain on convergence behavior. This perspective provides a conceptual bridge between optimization algorithms and feedback control, and establishes a foundation for the subsequent treatment of an optimizer as a dynamical system interconnected with the plant.

Presenter Biography

Yong Joo Do
Yong Joo Do received the B.S. degree in Mechanical and Aerospace Engineering from Seoul National University in 2024. Since 2024, he has been pursuing a combined M.S./Ph.D. program in Electrical and Computer Engineering at Seoul National University. His research interests include stochastic processes, multi-agent systems, and learning-based control.

Lecture 5

Soft-Constrained MPC: Feasibility and Stability

This talk introduces soft-constrained MPC for reference tracking as a systematic approach to mitigating infeasibility while retaining stability guarantees. State and terminal constraints are relaxed using slack variables, while hard input constraints are preserved and an artificial steady state is incorporated into the tracking formulation. These modifications enlarge the feasible region and permit controlled constraint violations when the corresponding hard-constrained problem is infeasible. The talk explains the roles of the slack penalties, terminal cost, terminal set, and artificial steady state, and presents the conditions used to establish asymptotic stability and input-to-state stability under additive disturbances.

Presenter Biography

Hyeonyeong Jang


Hyeonyeong Jang received the B.S. degree in Electrical Engineering from Hanyang University in 2024. Since 2024, he has been pursuing a combined M.S./Ph.D. program in Electrical and Computer Engineering at Seoul National University. His research interests include internal model principle, multi-agent systems, and learning-based control.

Lecture 6

Constraint Satisfaction under Inexact Optimization

This talk examines how constraint satisfaction can be maintained when real-time MPC operates with optimization error or experiences abrupt reference changes. An Explicit Reference Governor (ERG) generates an auxiliary reference from the optimization residual and available safety margin, thereby moderating command changes when the optimizer cannot track its solution rapidly. Dynamic Constraint Tightening (DCT), in contrast, reserves constraint margin to absorb the input error produced by a finite number of optimization iterations and updates the tightening level as the plant state changes. Finally, the talk explains how these updates balance performance, conservatism, recursive feasibility, and stability.

Presenter Biography

Hyungjo Byun
Hyungjo Byun received the B.S. degree in Electrical and Computer Engineering from the University of Seoul in 2023. Since 2023, he has been pursuing a combined M.S./Ph.D. program in Electrical and Computer Engineering at Seoul National University. His research interests include reinforcement learning and multi-agent systems.

Tutorial 4

Learning-Based Control for Automotive Systems: From Semi-Parametric Dynamics Modeling to Online Policy Improvement

OrganizerProf. Heejin Ahn (KAIST)
Date & TimeOctober 27 (Tue) 13:00-15:50
VenueHotel Emisia Sapporo Room T2 (Cattleya), 3F

Overview

TimeProgramSpeaker
13:00 – 13:10Opening & IntroductionOrganizer
13:10 – 14:25Learning-Based Modeling of Vehicle Dynamics:
Parametric, Non-parametric, and Semi-parametric Approaches
Prof. Cheolhyeon Kwon (UNIST)
14:25 – 14:35Break 
14:35 – 15:50Model Predictive Control as Online Policy Improvement:
A Reinforcement Learning Perspective
Prof. Kyunghwan Choi (KAIST)

Lecture 1

Learning-Based Modeling of Vehicle Dynamics:
Parametric, Non-parametric, and Semi-parametric Approaches
This tutorial covers different learning spectrums of vehicle dynamics modeling for trajectory-tracking control. We begin with parametric models, ranging from kinematic and dynamic bicycle models to higher-fidelity vehicle dynamics models, along with the offline and online techniques for identifying their parameters from data. We then turn to non-parametric models, principally Gaussian process regression, and their training pipelines and scalability limits. This motivates semi-parametric (grey-box) models, which augment a physics-based nominal model with a learned residual so that structure and data compensate for each other’s weaknesses. We review state-of-the-art methods trained both offline and through online iterative learning, demonstrated within the autonomous racing envelope.

Presenter Biography

Prof. Cheolhyeon Kwon
Cheolhyeon Kwon (Member, IEEE) received a B.S. degree in Aerospace Engineering from Seoul National University, Seoul, South Korea, in 2010, and an MS degree and Ph.D. from the School of Aeronautics and Astronautics, Purdue University, West Lafayette, IN, USA, in 2013 and 2017, respectively. He is currently an associate professor with the Department of Mechanical Engineering, UNIST, Ulsan, South Korea, and the director of the High-assurance Mobility Control laboratory. His research interests include control and estimation for dynamical cyber-physical systems (CPS), networked intelligent autonomy, and multi-agent system networks. His research goal is to pursue high assurance CPS design through the lens of control and estimation theory foundation, with applications to systems with mobility, such as autonomous driving, advanced air mobility, etc.

Lecture 2

Model Predictive Control as Online Policy Improvement:
A Reinforcement Learning Perspective
This tutorial presents model predictive control (MPC) as a practical framework for online policy improvement in learning-based control. We first revisit the relationship between optimal control, reinforcement learning, and policy improvement, and then show how an offline-trained policy can be refined online through finite-horizon MPC using model predictions, objective functions, and constraints. We further discuss how MPC-generated improved actions can provide training targets for updating an actor network, connecting online optimization with policy learning. The tutorial will combine theoretical insights with vehicle control examples to provide an intuitive understanding of MPC-based online policy improvement.

Presenter Biography

Prof. Kyunghwan Choi
Kyunghwan Choi received the B.S., M.S., and Ph.D. degrees in mechanical engineering from the Korea Advanced Institute of Science and Technology (KAIST), Daejeon, South Korea, in 2014, 2016, and 2020, respectively. He is currently an Assistant Professor with Cho Chun Shik Graduate School of Mobility, KAIST, and the Director of the Mobility Intelligence and Control Laboratory (MIC Lab). His research focuses on optimal and learning-based control for connected, automated, and electrified vehicles (CAEVs).

Tutorial 5

From AI & Robotics Research to Railway Deployment

OrganizerDr. Claire Nicodeme (ALSTOM)
Date & TimeOctober 27 (Tue) 14:00-17:00
VenueHotel Emisia Sapporo Room T3 (Azalea), 3F

Overview

TimeProgramSpeaker
14:00 – 14:15Opening & IntroductionOrganizer
14:15 – 14:35Railway Robotics Use Cases & System ArchitectureClaire NICODEME
14:35 – 15:05Perception in Railway RoboticsClaire NICODEME & Daisy CHAPMAN CHAMBERLAIN
15:05 – 15:15Break 
15:15 – 15:40Drones for Railway Inspection and MonitoringMatthieu LEVEQUE
15:40 – 16:05Telecommunications & Connectivity ConstraintsMarc FOURNIER
16:05 – 16:15Break 
16:15 – 16:30Ethics, Security & Human Factors in DeploymentClaire NICODEME & Marc FOURNIER & Daisy CHAPMAN CHAMBERLAIN
16:30 – 16:40From Research to Deployment – Lessons LearnedClaire NICODEME
16:40 – 17:00Q&AAll

Abstract

This tutorial bridges the gap between academic AI and robotics, and their application to industrial railway systems. It explains which robotic technologies are actually deployed in industry and how industrial constraints reshape algorithmic choices. A strong emphasis is placed on perception, highlighting its central role in task automation within harsh, repetitive, and evolving industrial environments, where robustness and human supervision are prioritized over raw performance.

Presenter Biography

Dr. Claire NICODEME
AI Project Owner (Alstom)
Industrial Artificial Intelligence & Robotics
Topics of recent interest include:
– Industrial Robotics,
– Railway Inspection & Automation,
– Perception Systems in Real-World Environments,
– AI Ethics.

Presenter Biography

Matthieu LEVEQUE
Head of the Mobilités Lab (SNCF VOYAGEURS) and Connectivity Business Development Engineer (e.SNCF Solutions)
Topics of recent interest include:
– Artificial Intelligence for Transportation
– Accessible Mobility and Inclusive Design
– Digital Transformation and Connectivity
– Living Labs and Open Innovation Ecosystems

Presenter Biography

Marc FOURNIER
Engineer
Innovation & Connectivity
Topics of recent interest include:
– Robotics Deployment & Connectivity (wired, wireless, latency),
– Communication Systems for Industrial Robotics